Geoscience Reference
In-Depth Information
The reinforcement strips at the upper layers were pulled out extensively from the
soil mass by the concrete panels that fell outward. At the lower layers, the
connections between the reinforcement strips and the concrete panels, which
were anchored into the concrete panel, were pulled out from the concrete panels.
This clearly illustrates the disadvantage of using rigid concrete panels as facing
material compared to flexible facing material such as geotextile. Hence, it can be
concluded the RS walls are more suitable for blast-resistant structures than RE
walls.
From the instrumentation program, the dynamic pressure recorded by the
total pressure cells installed in the RS wall showed that the peak incident pressure
in the soil, at a distance of about 3 m from the front of the wall, can be effectively
reduced to approximately 10% of its value at the front of the wall for all blast
events recorded. The instrumented data gave convincing proof of the efficient
dissipation of blast wave energy as it propagates into the depth of the RS wall,
making it a very efficient protective structure against blast loadings. Furthermore,
the RS wall can withstand a peak acceleration of about 20,000 g without any
deterioration. Hence, with the use of the geotextile reinforced soil wall, the blast
incident pressure can be reduced significantly and yet the wall was stable even
after several multiple blastings of similar intensity.
The dynamic response of the RS wall structure was studied using the new
PLAXIS Dynamics Module (version 7.2). Despite the difficulty of modeling the
exact details of the problem, a simplified 2D FEMmodel of the RS wall produced
results that matched reasonably well in trend with the field measurements of the
lateral stresses at two different locations of the RS wall, for the blast events MD5-
E1 and MD5-E2. With appropriate choice of soil dynamic and model parameters,
the FEM analysis clearly showed the importance of interface factors for the soil
response near the base of the wall. For the stress point close to the wall base, the
interface element plays a very crucial role to model realistic soil slippage
between the RS wall and the original ground, which is reflected in the matching
of the stress response for this location P (0.3m from wall base). Though exact
matching is not possible, the overall trend of stress increase and dissipation with
the blast loading is adequately shown in the calculations. For interface factors of
about 0.5 to 0.7, good agreement with measured response of soil pressures near
the base of the wall can be achieved. Thus dynamic FEM programs like PLAXIS
are capable of realistically modeling the dynamic response of reinforced soil wall
subjected to blast loading.
ACKNOWLEDGMENTS
The authors would like to acknowledge gratefully the support of LEO, Ministry
of Defence, and Singapore in this collaborative research and permission to
Search WWH ::
Custom Search